Buy new or revise and modify 30 year old class A amp and speakers?

[Raycoupe]

I'm having trouble finding experiences of people who have their old class A amps revised/modified. What to expect in sound quality?

I have a van Medevoort integrated class A amplifier MA220 I bought new in 1992. Due to circumstances it has been kept in storage 20 years and only used for 10 years. Revision and modification would cost nearly 600 euro's. Would it bring back original sound quality?

Since I do not own may other components at the moment, it's hard to determine the remaining sound quality of the amp. It's currently hooked up to a 25 year old harman kardon used as a pre-amp. It's sounds better than using the power amp of the Harman Kardon, but I realize it's pre-amp is probably not well suited for pairing with the van Medevoort. The lows are terribly abscent with the pre-amps bass setting on full blast.

The same goes for my Kef UniQ 90's speakers. Don't know how much soundquality is left, can't test it with a decent amp and don't know if it's worth replacing ferrofluid and revision of the cross over filter.

Perhaps soundquality has improved dramatically over 30 years and it's a now brainer to throw everything in the trash an buy new. But I simply don't know. Somehow I expect more progress is made on speakers than amps, but that is just a hunch.

I do now I was not very happy with the sound of a second hand 1,5 years old Yamaha r-n2000a, with a cheap S303 Yamaha CD-player and a pair of Dynaudio Evoke 50 speakers. It could be that the sellers room was the problem, not the equipment, but placement of instruments and vocals where all over the place. If this is what 30 years of devellopment has brought (although it's a class A/B amp) I would be very dissappointed.

Revision/mods of the MA220 would cost almost 600 euro's. Replacement 3000. (I know a 2 x 250 class D hypes ncore chinese amp would also cost 600, but that's a big gamble to buy without listening session)
Revision/mods of the Kef's would cost +/- 500. Replacement with i.e. Kef R3 Meta's 2200 euro's.

I could easily afford the revision/mods, buying new would be would be achievable, but not all at once.

 

[voodooless]

I think the restoration of the amp makes sense mostly for sentimental reasons. It's speced with a 92 SINAD, which is probably transparent, and when in good working condition, it will sound indistinguishable from any properly functioning modern amp, regardless of amp class. So don't be classist, the amp class alone tells you zero about the performance of an amp. The Medevoort looks stunning, though, so if it is worth it to you, go for it.

Amps and sources have no sound. They are generally transparent unless broken, and will all sound the same, despite what the industry would like you to believe. Speakers will make the biggest difference by far. Next is your room. The measurements of the Evoke don't look too bad, though. So it's likely the room was the issue here.

As for the KEF speaker, it's probably still quite good after revision. Why would the crossover need a revision? Generally, the caps used in those basically last forever.

Note that the R3 is a much smaller speaker. You may miss out on bass with those. So, possibly a sub might be a good idea.

If you want bigger, the Q11 might be of interest.

You can also look at fully integrated amps like WiiM Amp (Pro) at about €460.

 

[DVDdoug]

If it was me, I'd use the Yamaha, and I'm sure the CD player is fine too. But different speakers ALWAYS sound different (better or worse).

Revision and modification would cost nearly 600 euro's. Would it bring back original sound quality?

If the parts are available, I would assume so.

IMO, class-A is dumb. It's energy-inefficient and just not a good way to build a power amp. It's "A" because it was the first, NOT because it's best. In the tube days, radios & TVs were usually class-A because class A/B required another tube. And in the early transistor days, they could use one-less transistor. Now, the cost an additional transistor or MOSFET is insignificant. The cost per-watt is less with Class-A/B, and even lower with Class-D.

it's pre-amp is probably not well suited for pairing

You don't have to worry about "pairing" unless something is fouled-up.

The lows are terribly abscent with the pre-amps bass setting on full blast.

There's simply no excuse for that. Something is broken.

but that's a big gamble to buy without listening session)

It's usually not. Most electronics are better than human hearing unless you hear noise (hum, hiss, or whine in the background) or unless you overdrive an amplifier (or something) into clipping/distortion. Frequency response is usually flat across the audio range. (That's NOT the case with speakers/headphohnes. Most speakers don't have flat response and they all sound different.)

and don't know if it's worth replacing ferrofluid and revision of the cross over filter.

It should be obvious if the tweeter isn't working, or isn't working properly. And there is probably nothing wrong with the crossover. If something is wrong, and it's not as obvious as a dead tweeter, usually both speakers won't have the same exact failure/defect so you can compare them. But don't listen too carefully and "try" to hear a difference because speaker position makes a difference in sound and they can't both be in the same place at the same time, plus your brain will often imagine small differences.

but placement of instruments and vocals where all over the place.

That could be the speakers but it's NOT the electronics. It's a complicated illusion (with the sound actually coming from a pair of speakers).

Floyd Toole says:

The important localization and soundstage information is the responsibility of the recording engineer, not the loudspeaker.

But a dipole or omnidirectional speaker will make a difference, room acoustics make a difference, and of course your brain since it's an illusion.

Audiophoolery describes the few real characteristics of "sound quality" and it should help you ignore all of the nonsense you might read elsewhere.

 

[voodooless]

IMO, class-A is dumb. It's energy-inefficient and just not a good way to build a power amp

Yes, it is, but if you've checked out some pictures of the amp in question, you can see that the heatsinks would never accommodate a normal 100W Class A amp. This one is special in that it modulates the bias current with the signal, meaning that the heat loss is much smaller compared to a fixed bias Class A amp. Regardless, a modern class AB or D amp will be more efficient.

 

[Snarfie]

If it was me, I'd use the Yamaha, and I'm sure the CD player is fine too. But different speakers ALWAYS sound different (better or worse).


If the parts are available, I would assume so.

IMO, class-A is dumb. It's energy-inefficient and just not a good way to build a power amp. It's "A" because it was the first, NOT because it's best. In the tube days, radios & TVs were usually class-A because class A/B required another tube. And in the early transistor days, they could use one-less transistor. Now, the cost an additional transistor or MOSFET is insignificant. The cost per-watt is less with Class-A/B, and even lower with Class-D.


You don't have to worry about "pairing" unless something is fouled-up.


There's simply no excuse for that. Something is broken.


It's usually not. Most electronics are better than human hearing unless you hear noise (hum, hiss, or whine in the background) or unless you overdrive an amplifier (or something) into clipping/distortion. Frequency response is usually flat across the audio range. (That's NOT the case with speakers/headphohnes. Most speakers don't have flat response and they all sound different.)


It should be obvious if the tweeter isn't working, or isn't working properly. And there is probably nothing wrong with the crossover. If something is wrong, and it's not as obvious as a dead tweeter, usually both speakers won't have the same exact failure/defect so you can compare them. But don't listen too carefully and "try" to hear a difference because speaker position makes a difference in sound and they can't both be in the same place at the same time, plus your brain will often imagine small differences.


That could be the speakers but it's NOT the electronics. It's a complicated illusion (with the sound actually coming from a pair of speakers).

Floyd Toole says:

But a dipole or omnidirectional speaker will make a difference, room acoustics make a difference, and of course your brain since it's an illusion.

Audiophoolery describes the few real characteristics of "sound quality" and it should help you ignore all of the nonsense you might read elsewhere.

I would suggest starting with DSP room measurements before deciding on revisions or replacements. If you don't already have one, buy a measurement microphone DAC and connect it to a laptop or PC—Windows works well. Then use room correction software like REW (Room EQ Wizard) together with APO Equalizer or MathAudio Room EQ, in combination with Foobar2000. Both totaly free of cost.

The improvement can be dramatic. In my case, proper room correction made more than a 70% improvement in perceived sound quality. It’s a relatively low-cost step that helps you better understand what your current system is actually capable of before spending hundreds or thousands on upgrades or revisions.

 

[Raycoupe]

Great to here that revising speaker filters is probably a waste of money, since that is pretty expensive compared with replacing ferro cooling liquid.

I suspected that technology has advanced in 30 years, but I'm still not convinced enough that class D amplifiers with limited 500kHz switching power will sound great. In all measurements of such amps you'll see problems starting to appear when the audio frequency is above 1% of that 500kHz, like in the test reviews here.

I choose that van Medevoort 33 years ago because I had access to an expensive spectrum analyzer from my university and the van Medevoort was phase correct all the way across the spectrum and most important, between both channels. And it sounded like a whole new level compared to equally expensive A/B amps. As far as I have experienced with a couple of amps in that era, Quad, Accuphase, Krell and some expensive A/B amps, that's what delivers exact placement of voices and instruments. Not very noticeable playing popular music, but on good jazz recordings with a more empty sound stage like ECM label it was very noticeable.

But that was more than 30 years ago, and I don't have any experience with current class A/B en class D amps. And I really don't care much what technique is used, less heat is very welcome, in the end the ears will decide what's best, not the specs.

It sounds very appealing a 2 x 250 watt poweramp for 600 euro's but there are no physical shops where such amplifiers can be heared. I can only order one online.

On bass performance: I have neighbours nowadays and the same low defined bass like it had 30 years ago, is not wanted anymore.

 

[SSS]

I'm having trouble finding experiences of people who have their old class A amps revised/modified. What to expect in sound quality?

I have a van Medevoort integrated class A amplifier MA220 I bought new in 1992. Due to circumstances it has been kept in storage 20 years and only used for 10 years. Revision and modification would cost nearly 600 euro's. Would it bring back original sound quality?

Since I do not own may other components at the moment, it's hard to determine the remaining sound quality of the amp. It's currently hooked up to a 25 year old harman kardon used as a pre-amp. It's sounds better than using the power amp of the Harman Kardon, but I realize it's pre-amp is probably not well suited for pairing with the van Medevoort. The lows are terribly abscent with the pre-amps bass setting on full blast.

The same goes for my Kef UniQ 90's speakers. Don't know how much soundquality is left, can't test it with a decent amp and don't know if it's worth replacing ferrofluid and revision of the cross over filter.

Perhaps soundquality has improved dramatically over 30 years and it's a now brainer to throw everything in the trash an buy new. But I simply don't know. Somehow I expect more progress is made on speakers than amps, but that is just a hunch.

I do now I was not very happy with the sound of a second hand 1,5 years old Yamaha r-n2000a, with a cheap S303 Yamaha CD-player and a pair of Dynaudio Evoke 50 speakers. It could be that the sellers room was the problem, not the equipment, but placement of instruments and vocals where all over the place. If this is what 30 years of devellopment has brought (although it's a class A/B amp) I would be very dissappointed.

Revision/mods of the MA220 would cost almost 600 euro's. Replacement 3000. (I know a 2 x 250 class D hypes ncore chinese amp would also cost 600, but that's a big gamble to buy without listening session)
Revision/mods of the Kef's would cost +/- 500. Replacement with i.e. Kef R3 Meta's 2200 euro's.

I could easily afford the revision/mods, buying new would be would be achievable, but not all at once.

Myself I would keep this good amplifier. The data seem to be good enough for good sound. And, the amp is not load dependable. Why want you to revise this amp.
I have amps never revised from the 1980ies and these work still without problems.

 

[voodooless]

In all measurements of such amps you'll see problems starting to appear when the audio frequency is above 1% of that 500kHz, like in the test reviews here.

What problems?

 

[Raycoupe]

Yes, it is, but if you've checked out some pictures of the amp in question, you can see that the heatsinks would never accommodate a normal 100W Class A amp. This one is special in that it modulates the bias current with the signal, meaning that the heat loss is much smaller compared to a fixed bias Class A amp. Regardless, a modern class AB or D amp will be more efficient.

Yes, dynamic class A. In stead of 220 watts of power it will only produce 100 watts of heat when idle, 50 watts per side. It will keep a current for 50 watts of dynamic range per channel, so when you use 2 x 30 watts the heat output will be 2 x (30 +50) = 160 watts and it will get pretty hot. Using more power and you could keep a two pot of tea hot indefinitely.

To use up to 60 continuous watts per channel on this amp, requires an upgraded power cord, but realistically 60 watts will be deafening levels of sound in a living room.

And that much heat will kill most electronic devices you put on top of the amp! It should stand alone, with plenty of room around the damn thing. I didn't do that with my Teac VRDS 10 cd player so it died early. I definitely could do with less heat.

Thank you all for responding!

Myself I would keep this good amplifier. The data seem to be good enough for good sound. And, the amp is not load dependable. Why want you to revise this amp.
I have amps never revised from the 1980ies and these work still without problems.

I simply don't know how good the amp still is. It it the source, the harman kardon used a pre-amp, the van Medevoort or the speakers, all could be the reason I get lower treble and bass. The Harman Kardon is only there because it has a bass treble adjustment to crank things up.

 

[Raycoupe]

What problems?

View attachment 469257

I.e this review https://www.audiosciencereview.com/forum/index.php?threads/hypex-nilai500diy-amplifier-review.41669/
See:

Above 5 kHz (1% of the power switching freq) things go south.

And the weird spike around 10kHz:

And how could a 20kHz signal be amplified correctly with a power switching only 25 times of that frequency? That's very low resolution so to me it is no surprise that it will struggle at high frequencies. But hey, that whole class D amplification is new to me, I feel like a noob again.

These tests are all about noise, distortion, wattage and frequency response and are already far from perfect. But what about the phase accuracy of the signal?

I know that in home multi channels surround sound systems and PA systems blasting away at concerts and clubs, phase correctness is of no concern. Varying distances and offset positions of the listener already throws that out the door. That's why we use 5.1 systems at home when viewing movies with family and friends, so all can experience 3D sound. Any analog bass/treble adjustment or equalizer and phase accuracy will be down the drain as well.

But for high end audio, sitting in the sweetspot close to 2 speakers, it's pretty important. Why isn't that tested anymore? Needed spectrum analyzer equipment doesn't cost a million bucks like 30 years ago.

Have things changed and are all amps phase accurate/correct nowadays?

In the end, no specifications or figures matter if the ears don't agree. I would never buy any audio equipment based on specs and test results alone.

In our hifi shop back then, we had a big screen with soundwave transparent cloth (same as on speakers) to obscure the equipment from the client. Client were almost always very biased and came in looking for Accuphase equipment (there's that phase again, the name is derived from accurate and phase). Only by listening blindly to other equipment we were able to convince them that sometimes other setups could sound a lot better for their taste, for less money.

 

[voodooless]

I.e this review https://www.audiosciencereview.com/forum/index.php?threads/hypex-nilai500diy-amplifier-review.41669/
See:

Above 5 kHz (1% of the power switching freq) things go south.

You cannot hear any of that. What you can hear is in the multitone plot.

And how could a 20kHz signal be amplified correctly with a power switching only 25 times of that frequency? That's very low resolution so to me it is no surprise that it will struggle at high frequencies. But hey, that whole class D amplification is new to me, I feel like a noob again.

Good luck passing a double blind test. You won’t.

These tests are all about noise, distortion, wattage and frequency response and are already far from perfect. But what about the phase accuracy of the signal?

These electronics are minimum phase, meaning phase can be extracted from frequency response. If the response is flat, so is the phase. Class D amps will have a slight phase shift at the very end due to their lower bandwidth, but that is inaudible:

'Does Phase Distortion/Shift Matter in Audio? (no*)'

Jun 7, 2021

A viewer of my videos and member he suggested that I do a video commenting on a video that Paul McGown did on audibility of phase shifts. Here is Paul's video which was really about a different question (why we need wideband amplifiers) but turned it into phase being an audible problem:

Here is my answer to him:

Of course phase is an important electrical and acoustic thing and has relevance in countless situations. It is just that it should not be used to create myths and fear in audiophiles with respect to audibility in the context Paul and others...

• amirm
• Replies: 1,341
• Forum: Audio Science Review Video Channel

• 

But for high end audio, sitting in the sweetspot close to 2 speakers, it's pretty important. Why isn't that tested anymore? Needed spectrum analyzer equipment doesn't cost a million bucks like 30 years ago.

Because it’s pointless.

Have things changed and are all amps phase accurate/correct nowadays?

They probably always have been, unless they were broken.

In the end, no specifications or figures matter if the ears don't agree. I would never buy any audio equipment based on specs and test results alone.

Then at least let ONLY your ears decide:

In our hifi shop back then, we had a big screen with soundwave transparent cloth (same as on speakers) to obscure the equipment from the client. Client were almost always very biased and came in looking for Accuphase equipment (there's that phase again, the name is derived from accurate and phase). Only by listening blindly to other equipment we were able to convince them that sometimes other setups could sound a lot better for their taste, for less money.

I’ll refer you again to the YouTube link above.

 

[Raycoupe]

@voodooless I will look into those video's tomorrow, getting late her, and yes my ears will decide eventually. But you know with youtube vids, ones saying the opposite are just as easy to find, with a ton of other vids on the right claiming also that phase shift matters:

If the audio source, the studio recording and producer don't care, it will certainly not matter. And I will be fair, this is the case for most productions. That's why people seek out labels like ECM, Blue Note and Deutsche Grammophon.

Like I said in another thread, I will start by having my ears tested. I'm 53 now, not 23. I want to know before I bug hifi stores to demonstrate equipment.

And those cheap amps, with those specs and test results are having my attention. It might even be worth trying. What has 30 years of devellopment brought to the table. I could always resell at half price and still learn a valuable lesson.

What hasn't changed in 30 years is that people will not always agree and hearing is very subjective.

 

[antcollinet]

how could a 20kHz signal be amplified correctly with a power switching only 25 times of that frequency?

How can a 20kHz signal be perfectly reproduced with only two samples per period (Nyquist). 25 pulses per period is massive by comparison.

 

[Sokel]

Revise, 100% .

 

[Raycoupe]

How can a 20kHz signal be perfectly reproduced with only two samples per period (Nyquist). 25 pulses per period is massive by comparison.

Very valid argument for cd quality. Common sampling frequency nowadays for hi-res flac, etc is 192 kHz, still not 25 samples at 20 kHz. Higher sample rates PCM (up to 786) are available and needed. But what about DSD.

Bitstream DSD (DSD512 has a sample rate of 22.5792MHz) works completely different of course, but is already equal to 24 bit 192 in sound quality with much higher resolution but a bit less dynamic range. A whole debate is going on which sounds better for high end use.

Although the amplitude representation in 24 and /32 bit PCM is better, but DSD512 resolution is vastly superior. 44 nano seconds vs 5200 ns for 24/192.

So then 25 amp samples suddenly look poor when that 20kHz frequency is sampled 2000+ times! (and still 250+ times for DSD64)

DSD1024 is coming and already incorparated in pricy new hardware and will double that again, also storing ten times the information of 24 bit/192.

I can't find alll my favorite albums yet in DSD512, but many are already available. I guess it will remain a mix of PCM and DSD and some MQA (compressed format), depending on availability and price!

File size doesn't matter much anymore. Storage is cheap, I don't count in Gb's but Tb's.

 

[voodooless]

Very valid argument for cd quality. Common sampling frequency nowadays for hi-res flac, etc is 192 kHz, still not 25 samples at 20 kHz.

A 20 kHz sine isn’t better reconstructed with 192 kHz vs 44.1 kHz sampling. If you don’t understand that, you don’t understand the sampling theorem.

Higher sample rates PCM (up to 786) are available and needed.

They aren’t needed You hear until about 20 kHz (if you’re lucky), so 44.1 kHz is enough for everyone.

Bitstream DSD (DSD512 has a sample rate of 22.5792MHz) works completely different of course, but is already equal to 24 bit 192 in sound quality with much higher resolution but a bit less dynamic range. A whole debate is going on which sounds better for high end use.

Stop drinking the marketing cool aid the industry has been feeding you for decades. You don’t need any of this. It audibly doesn’t bring any benefits, and countless studies have shown this.

Although the amplitude representation in 24 and /32 bit PCM is better, but DSD512 resolution is vastly superior. 44 nano seconds vs 5200 ns for 24/192.

Again, you don’t understand the sampling theorem… sample rate has little to do with time resolution.

'Time resolution of Redbook (16/44) PCM'

Apr 4, 2021

Ok, that old BS about "only one sample time resolution" came up again elsewhere. I made the following demonstration to put the screws to that.

I generated a 10kHz sine wave at 44.1 kHz sampling rate. The matlab file is attached. I generated 128 discrete phases uniformly split over one cycle, so each new phase shift is equal to 4.41/128 sample delay, or about .7 microseconds. The routine plots the first 8, making the very obvious point that they are different. Q. (obscenity) E. D.

Note, I had to add a .txt to the filename to get ASR to load it, apparently we don't like .m files. If you...

• j_j
• Replies: 198
• Forum: Psychoacoustics: Science of How We Hear

• 

DSD1024 is coming and already incorparated in pricy new hardware and will double that again, also storing ten times the information of 24 bit/192.

More information isn’t stored in the audible spectrum. I again refer to the sampling theorem.

I can't find alll my favorite albums yet in DSD512, but many are already available. I guess it will remain a mix of PCM and DSD and some MQA (compressed format), depending on availability and price!

File size doesn't matter much anymore. Storage is cheap, I don't count in Gb's but Tb's.

Don’t bother, stick with what you have, it’s more than good enough.

And again we’re wasting time on silly electronics, while 90% of the sound quality is determined by the speaker and the room interactions. Focus on that, it will be much more rewarding and will actually make a massive difference.

 

[antcollinet]

Very valid argument for cd quality.

And yet it isn't. A 20kHz sine wave - or any signal of 20Khz bandwidth can be perfectly reconstructed from 44.1Khz sampling. I was just about to post the monty video - but see @voodooless has already done it.

If you watch a few times (it probably took me three viewings to get my head around it the first time I saw it - it takes time to throw out ones misconceptions) you will probably realise that many of your ideas about the quality of digital audio are based on misunderstandings. Especially about time resolution, and phase issues (see from 20:50 onwards)

Here is another article showing how the human autitory system cannot hear the any improvement from anything better than 16/44.1

24/192 Music Downloads are Very Silly Indeed

people.xiph.org

 

[JSmith]

But what about DSD.

There is nothing good or advantageous about DSD... unless you like large files with high frequency noise taking up space.


JSmith

 

[antcollinet]

And yet it isn't. A 20kHz sine wave - or any signal of 20Khz bandwidth can be perfectly reconstructed from 44.1Khz sampling. I was just about to post the monty video - but see @voodooless has already done it.

If you watch a few times (it probably took me three viewings to get my head around it the first time I saw it - it takes time to throw out ones misconceptions) you will probably realise that many of your ideas about the quality of digital audio are based on misunderstandings. Especially about time resolution, and phase issues (see from 20:50 onwards)

Here is another article showing how the human autitory system cannot hear the any improvement from anything better than 16/44.1

24/192 Music Downloads are Very Silly Indeed

people.xiph.org

@Raycoupe Here is an analogy that helped me.

Someone wants to communicate to you how to draw a circle on a piece of paper. They can send you two pieces of information. The coordinates of the centre and the diameter of the circle. They don't need to send you the coordinates of a million points around the edge of the circle.

BECAUSE YOU KNOW THAT THEY ARE DESCRIBING A CIRCLE - you can draw a perfect representation of what they want from just those two pieces of information.


What the sampling theorem proves is that you can perfectly reconstruct a band-limited signal from samples at 2x the maximum frequency in the signal. This only works BECAUSE YOU KNOW YOU ARE RECONSTRUCTING A BAND-LIMITED SIGNAL.

Just as you can draw a circle from two data points because you know the shape you are reconstructing - so you can reconstruct a sine wave (or any signal containing sine waves no longer than two sample periods) with only two data points per cycle - because you know the shape you are reconstructing.

 

[mhardy6647]

Here is another article showing how the human [auditory] system cannot hear the any improvement from anything better than 16/44.1

But it should always be borne in mind in these discussions that we're not talking about humans -- we're talking about audiophiles.

There is nothing good or advantageous about DSD... unless you like large files with high frequency noise taking up space.

So very American (and I say this as an American )... the RAM 2500 of audio data files.